# Learning to Predict Robot Keypoints Using Artificially Generated Images

## TL;DR

This paper presents a method for robot keypoint estimation using artificially generated images with a feedback mechanism that adapts rendering probabilities, achieving high accuracy and reducing training steps.

## Contribution

Introduces a feedback-based probabilistic rendering approach for supervised robot keypoint estimation, improving training efficiency and accuracy.

## Key findings

- Achieves near-human accuracy on real images.
- Reduces training steps needed for model convergence.
- Maintains model quality with synthetic data.

## Abstract

This work considers robot keypoint estimation on color images as a supervised machine learning task. We propose the use of probabilistically created renderings to overcome the lack of labeled real images. Rather than sampling from stationary distributions, our approach introduces a feedback mechanism that constantly adapts probability distributions according to current training progress. Initial results show, our approach achieves near-human-level accuracy on real images. Additionally, we demonstrate that feedback leads to fewer required training steps, while maintaining the same model quality on synthetic data sets.